The Diamond Store is a leading online retailer of fine jewelry in the United Kingdom. The business was profitable with their existing marketing strategy, but they found it difficult to attribute sales to their campaigns accurately. This prevented them from understanding which campaigns were effectively driving sales and revenue – and which ones weren’t.
The business also experienced a longer than= average sales cycle, making it even more difficult to track the impact of their marketing on their bottom line.
They turned to Omnitail to help them solve these attribution problems and grow sales and revenue profitably!
2x
Increase in Orders
Year/Year
2.3x
Increase in Revenue
Year/Year
3.6x
Increase in Profit
Year/Year
INVESTIGATING
THE ACCOUNT
The Diamond Store had profitable campaigns when we took over the account, but there were a few problems with their existing approach.
We needed to tackle two main issues.
1. Problems with Attribution
The Diamond Store did not have a way to precisely measure the impact of their marketing efforts. The business itself was profitable – but they had no way of knowing which campaigns were creating profit, or how much they should be pending to maximize it.
2. Long Sales Cycles
Jewelry purchases require a lot of deliberation on the part of the consumer. This leads to a long sales cycle that often involves multiple touchpoints. This issue exacerbated the above attribution problems because the usual last-click model couldn’t accurately assign credit for sales to campaigns.
The chart above visualizes the number of times customers interact with The Diamond Store’s marketing before converting. As you can see, the vast majority require more than two touchpoints, and a significant margin require more than five or even 10+. These additional touchpoints indicate a much longer sales cycle.
Omnitail's Advertising Strategy
Multi-touch Attribution
As we saw above, the purchase cycle for the Diamond Store’s products was often long and contained multiple touchpoints. Last-click attribution, though, ignored everything except the last touchpoint prior to the sale, assigning all credit to the last interaction. For the Diamond Store, this meant that branded campaigns were far overrepresented in their metrics. Worse, the model gave no credit to PLAs that introduced customers to the brand, as they typically appeared early in the buyer journey.
We began to measure performance according to multiple attribution models. For example, we introduced an assisted + last-click model, which gives credit to any and all campaigns that touched the buyer along the journey. PLAs are represented in our metrics even if the customer interacted with them early in the buying cycle. This gives us a much more accurate picture of the touchpoints that are influencing sales.
Data-Driven Spend Allocation
There was one more step to sorting out attribution: choosing the correct lookback window. A lookback window is the time after an ad is clicked in which potential sales can be attributed to that click. Longer lookback windows (i.e., 30-60 days) can better account for the additional time it takes the Diamond Store’s customers to complete the buyer journey. This prevents underrepresenting campaigns whose touchpoints take place early in the buying cycle.
RESULTS
Prior to working with Omnitail, the Diamond Store’s campaigns were profitable. However, attribution problems and a long sales cycle made allocating spend efficiently a difficult task.
Omnitail helped them correctly attribute their sales to their campaigns. Implementing a multi touch attribution model meant we could also measure ad interactions that were previously not being represented. Additionally, we began to evaluate revenue and profit using a longer lookback window. This gave us a much better idea of the campaigns that were influencing sales. With these building blocks in place, it became much easier to spend effectively and grow profit!
